Skip to content
BranchesTags

Latest commit

 

History

17 Commits

Folders and files

NameName
Last commit message
Last commit date
.playwright-mcp
.playwright-mcp
 
 
checkpoints
checkpoints
 
 
data
data
 
 
kaggle_benchmarks
kaggle_benchmarks
 
 
logs
logs
 
 
markdowns
markdowns
 
 
orbit_wars
orbit_wars
 
 
runs
runs
 
 
tests
tests
 
 
.codex
.codex
 
 
.env.example
.env.example
 
 
.gitignore
.gitignore
 
 
AGENTS.md
AGENTS.md
 
 
CLAUDE.md
CLAUDE.md
 
 
GPU-CPU-plan.md
GPU-CPU-plan.md
 
 
KaggleBriefer.md
KaggleBriefer.md
 
 
LONG_TERM_FIXES.md
LONG_TERM_FIXES.md
 
 
action_space.py
action_space.py
 
 
agents.py
agents.py
 
 
agents_cpu.py
agents_cpu.py
 
 
bc_data.py
bc_data.py
 
 
bc_data_cpu.py
bc_data_cpu.py
 
 
bc_train.py
bc_train.py
 
 
bc_train_chunked_cpu.py
bc_train_chunked_cpu.py
 
 
bc_train_cpu.py
bc_train_cpu.py
 
 
bench.py
bench.py
 
 
bench_cpu.py
bench_cpu.py
 
 
chunked_bc_cpu.py
chunked_bc_cpu.py
 
 
chunked_model_cpu.py
chunked_model_cpu.py
 
 
eval_bc.py
eval_bc.py
 
 
eval_bc_cpu.py
eval_bc_cpu.py
 
 
getting-started.ipynb
getting-started.ipynb
 
 
harness.py
harness.py
 
 
harness_cpu.py
harness_cpu.py
 
 
make_chunked_submission.py
make_chunked_submission.py
 
 
make_cpu_submission.py
make_cpu_submission.py
 
 
make_nn_submission.py
make_nn_submission.py
 
 
make_submission.py
make_submission.py
 
 
model.py
model.py
 
 
model_cpu.py
model_cpu.py
 
 
radar.py
radar.py
 
 
rl_opponent_pool.py
rl_opponent_pool.py
 
 
rl_opponent_pool_cpu.py
rl_opponent_pool_cpu.py
 
 
rl_ppo.py
rl_ppo.py
 
 
rl_ppo_cpu.py
rl_ppo_cpu.py
 
 
rl_rollout.py
rl_rollout.py
 
 
rl_rollout_cpu.py
rl_rollout_cpu.py
 
 
rl_train.py
rl_train.py
 
 
rl_train_chunked_cpu.py
rl_train_chunked_cpu.py
 
 
rl_train_cpu.py
rl_train_cpu.py
 
 
server.py
server.py
 
 
server_model_eyes.py
server_model_eyes.py
 
 
submission.py
submission.py
 
 
submission_chunked.py
submission_chunked.py
 
 
submission_chunked_m2.py
submission_chunked_m2.py
 
 
submission_cpu.py
submission_cpu.py
 
 
submission_nn.tar.gz
submission_nn.tar.gz
 
 
submission_nn.zip
submission_nn.zip
 
 
targeting.py
targeting.py
 
 

About

RL self-play agent for a 1-second-per-turn CPU-budget Kaggle game. 100× inference speedup via radar + dynamic tokens + axial attention.

www.kaggle.com/competitions/orbit-wars/overview
Activity

Stars

1 star

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages